. 2011 Jun 10;6(6):e20943. doi: 10.1371/journal.pone.0020943

Table 1. Dependence of the surface parameters of the NCD samples on the boron doping level.

Sample Type	B Doping Level	Roughnessrms	Surface Potential^*	AFM phase rms	Resistivity	Contact Angle
I. 0 ppm	N/A	27±3 nm^{II, IV}	40±4 mV^{II, III, IV}	2±1 deg^II	>10 MΩ	17±6°^IV
II. 133 ppm	2×10¹⁹ cm⁻³	20±3 nm^{I, III, IV}	96±6 mV^{I, III, IV}	4±1 deg^I	55 kΩ	18±5°^IV
III. 1000 ppm	1×10²⁰ cm⁻³	27±4 nm^{II, IV}	47±4 mV^{I, II, IV}	3±1 deg	0.6 kΩ	20±5°^IV
IV. 6700 ppm	6×10²⁰ cm⁻³	33±4 nm^{I, II, III}	42±5 mV^{I, II, III}	3±1 deg	0.3 kΩ	14±4°^{I, II, III}

*Values adjusted to the surface potential of gold.

For the NDP measurements, the accuracy can achieve 5% (which is the precision of the boron atoms in the etalon). However, the precision of the NDP technique also depends on other parameters, e.g. the stability of the neutron beam intensity, identical geometry of the etalon and the measured sample, etc. Realistically, the NDP data can be routinely measured with accuracy of 10% in our case.

For roughness, potential, phase and contact angle, the data is presented as Mean ± S.D. (Standard Deviation). In the case of roughness and AFM phase, each rms value was determined from 65 536 data points on each sample type. The mean and S.D. of rms values were calculated from 5 such measurements across the sample. In the case of surface potential, the mean and S.D. values were calculated from 65 536 measurements across each sample. The contact angle was calculated from fitting the curve of the water droplet, and the mean and S.D. values were calculated from 16 measurements for each sample type.

Statistical Analysis: ANOVA, Student-Newman-Keuls Method. Statistical significance: ^{I, II, III, IV}: p≤0.05 compared to the group labelled with the same Roman number.

For the room temperature electrical resistivity measurements, the accuracy is better than 1%.